The conventional approach to an engineering design and development effort tends to be costly and cumbersome. Typically, a design and development effort begins with an initial idea or concept for a new product, for example, an improved sonobuoy for receiving and transmitting to a surveiling aircraft acoustic data from the ocean.
From this initial idea, an initial design comprising drawings and written specifications is created. The drawings may be paper drawings or “blueprints” or may be three dimensional drawings created using Computer Aided Design (CAD) software. The initial design will usually include certain specifications, such as product specifications (defining performance of the item), system specifications (defining performance of a system including the item) and interface design specifications (defining electronic or physical interfaces between the item and other system components). In the case of a sonobuoy, these documents will include a product specification detailing the performance criteria and “form, fit and function” of the sonobuoy; a system specification detailing the performance of the system including the sonobuoy and aircraft avionics; and an interface design specification defining the electronic interface between the sonobuoy and the aircraft receiver and specialized acoustical processors.
These specifications will include various design criteria or performance parameters for the new sonobuoy, such as reliability requirements, mean time between failure, radio frequency (RF) transmission power and range, acoustic gain and performance (e.g., gain in decibels and three dimensional beam patterns) and so on.
As depicted in FIG. 1, this initial idea or concept is translated into an initial design, “Design 1.” A series of teams, comprising one or more members, are assembled to evaluate the design according to their various engineering, business and management specialties. Typically, there will be a management team whose primary function is to ensure that the development program is “on schedule” and “on cost.” There will be a business/accounting team who will analyze the costs of the development program and the expected costs of the product produced in quantity. There will be a series of engineering teams, each of a particular discipline or specialty. In our sonobuoy example, there will be mechanical engineers, electrical engineers, RF engineers, acoustic specialists, reliability engineers, safety engineers, signal processing specialists, production engineers and so on.
Each of these teams will evaluate the design (three dimensional model and associated specifications) to determine conformance with requirements. In a typical development effort, these teams will recommend changes to the design to rectify deficiencies or improve performance. The design will evolve (as depicted in FIG. 1, from Design 1, Design 2, and so on to final Design n) as changes are recommended and implemented.
Because a number of disciplines may be involved, it can be appreciated that the process can be lengthy and costly. When one team implements changes based on its analysis, other teams may have to redo their analysis as a result. For example, when a production engineer recommends (based on cost or manufacturibility considerations) that paper capacitors be employed instead of ceramic capacitors, the reliability engineer will need to reevaluate the design. When an acoustic specialist recommends that improved hydrophones be used or an RF specialist recommends that an improved receiver be used, the mechanical engineer will need to determine if packaging limits are exceeded and the accounting specialists will need to determine if “design to cost” parameters are exceeded.
Application of the “concurrent engineering” principle can improve the process somewhat by assembling the various specialists at the earliest possible time. Thus, manufacturing specialists are consulted from the beginning, rather than simply at the end.
Nevertheless, the sonobuoy example and FIG. 1 illustrate how a complex engineering effort involving multiple disciplines can be a lengthy and costly process as designs are considered and discarded, new designs are evaluated and so forth. When the various specialists are in different geographical locations (“physical boundaries”), at different business entities (“business boundaries”), or employ different software standards (“format boundaries”), the problem is only compounded. The disparate locations, business cultures and format standards can be significant impediments. This can be referred to as the “boundaries problem.” Put simply, having specialists of differing disciplines (with often conflicting priorities) attempt to resolve design issues when operating from different locations, or from business entities with differing business practices and cultures, or when using different software formats (e.g., three dimensional model standards) creates significant costs and obstacles. This is a significant drawback.
Once an acceptable design is arrived at (e.g., Design n, FIG. 1), the developing concern typically wishes to produce the design in quantity. In today's decentralized economy, where most “start-up” or even moderately-sized engineering enterprises do not have their own production facilities, this typically requires “outsourcing” of the production. In some respects, the process of locating qualified, performance-proven fabricators can be as daunting as the engineering development effort. The engineering concern may not have established relationships with many—or any—manufacturers. Of the few potential candidates that may be identified through “word of mouth” or a costly search, it can be costly to evaluate whether such candidates meet minimum requirements. Numerous meetings may be required. Even then, it may be difficult to ascertain the quality of past performance and the prospects for the proposed performance. The engineering concern will have to sustain the cost, and risk, of divulging proprietary design and specification data. Because format standards may differ, the costs and risks of (sometimes imperfect) file conversions may be required. These are significant drawbacks.
Even if several acceptable fabricators are located, they may represent but a fraction of the otherwise qualified pool of fabricators. This lessens competition and ultimately, can lead to increased costs and decreased performance. These are significant disadvantages to the engineering concern. With respect to start-up or not-well-known fabricators, this is a significant drawback that prevents them from penetrating new markets and inhibits their growth.
Once a pool (or even a single) of qualified fabricators is identified, the process of negotiating an agreement on performance must occur. This may require numerous phonecalls, teleconference calls, and face-to-face meetings. Three dimensional models and specification documents may have to be divulged so that the fabricator can develop a bid. Despite rigorous concurrent engineering, it is common that minor and not-so-minor redesigns may be required due to producibility concerns. This will require an additional engineering effort between the supplier and the proposed fabricator to discuss and decide on engineering changes to arrive at a mutually acceptable, and producible, design that meets and performance and cost requirements. The time and effort entailed are significant drawbacks.
Engineering issues aside, the negotiation process can be lengthy and costly. Documents, which may be proprietary, are sent to the bidding fabricator. The bidding fabricator must evaluate them, engage in discussions with the prime contractor about the design and specifications, and generate a “bid.” This bid typically includes terms relating to time, cost and performance. Upon receipt of the bid, additional discussions between the parties may be required to address engineering and/or contractual issues. Typically conducted using a combination of the phone, teleconference and postal service (or the like), these discussions impose additional costs and time in getting a product to market. This is a significant drawback.
Other problems and drawbacks also exist.